Download 4

Bioconjugate Chem. 2007, 18, 894−902
894
The Same Drug but a Different Mechanism of Action: Comparison of Free
Doxorubicin with Two Different N-(2-Hydroxypropyl)methacrylamide
Copolymer-Bound Doxorubicin Conjugates in EL-4 Cancer Cell Line
Lubomı́r Kovář,*,† Jiřı́ Strohalm,‡ Petr Chytil,‡ Tomáš Mrkvan,† Marek Kovář,† Ondřej Hovorka,†
Karel Ulbrich,‡ and Blanka Řı́hová†
Department of Immunology and Gnotobiology, Institute of Microbiology ASCR, v.v.i., Vı́deňská 1083,
142 20 Prague 4, Czech Republic, and Department of Biomedical Polymers, Institute of Macromolecular Chemistry
ASCR, v.v.i., Heyrovský Square 2, 162 06 Prague 6, Czech Republic. Received August 8, 2006;
Revised Manuscript Received February 8, 2007
Doxorubicin is one of the most potent anti-tumor drugs with a broad spectrum of use. To reduce its toxic effect
and improve its pharmacokinetics, we conjugated it to an HPMA copolymer carrier that enhances its passive
accumulation within solid tumors via the EPR effect and decreases its cytotoxicity to normal, noncancer cells. In
this study, we compared the antiproliferative, pro-survival, and death signals triggered in EL-4 cancer cells exposed
to free doxorubicin and doxorubicin conjugated to a HPMA copolymer carrier via either enzymatically (PK1) or
hydrolytically (HYD) degradable bonds. We have previously shown that the intracellular distribution of free
doxorubicin, HYD, and PK1 is markedly different. Here, we demonstrated that these three agents greatly differ
also in the antiproliferative effect and cell death signals they trigger. JNK phosphorylation sharply increased in
cells treated with HYD, while treatment with free doxorubicin moderately decreased and treatment with PK1
even strongly decreased it. On the other hand, treatment with free doxorubicin greatly increased p38 phosphorylation,
while PK1 and HYD increased it slightly. PK1 also significantly increased ERK phosphorylation, while both the
free doxorubicin and HYD conjugate slightly decreased it. Long-term inhibition of JNK significantly increased
both proliferation and viability of EL-4 cells treated with free doxorubicin, showing that the JNK signaling pathway
could be critical for mediating cell death in EL-4 cells exposed to free doxorubicin. Both activation of caspase
3 and decreased binding activity of the p50 subunit of NFκB were observed in cells treated with free doxorubicin
and HYD, while no such effects were seen in cells incubated with PK1. Analysis of the expression of genes
involved in apoptosis and regulation of the cell cycle demonstrated that free doxorubicin and HYD have very
similar mechanisms of action, while PK1 has very different characteristics.
INTRODUCTION
Doxorubicin is one of the most effective anthracycline
antibiotics with a broad anti-tumor spectrum. Unfortunately,
numerous side effects such as severe cardiotoxicity, bone
marrow suppression, and alopecia limit its use. Development
of tumor-targeted formulations is one of the main lines of
research for improving the safety and efficacy of anthracyclines.
For this purpose, we have used a water-soluble, nontoxic
copolymerbasedonN-(2-hydroxypropyl)methacrylamide(HPMA)
as a carrier for anti-cancer drugs. Covalent conjugation of
doxorubicin to HPMA copolymer eliminates its adverse cytotoxicity, as this conjugate is biologically inactive when circulating in the bloodstream (1). Enhanced and tumor-selective
accumulation of HPMA conjugate within the solid tumor is
achieved either passively due to the enhanced permeability and
retention (EPR) effect (2) or actively by addition of a targeting
moiety (e.g., antibodies, lectins, carbohydrates) (3).
Anthracyclines cause apoptosis of cancer cells by a complex
network of events which include intercalation into the DNA,
generation of reactive oxygen species, topoisomerase II inhibition, and DNA damage (4). HPMA conjugate that contains
doxorubicin bound via a pH-sensitive bond (HYD) possesses
similar biological properties as free doxorubicin (5), because
* Corresponding author. Tel.: +420241062158. Fax: +420241721143.
E-mail address: lkovar@biomed.cas.cz.
†
Institute of Microbiology ASCR.
‡
Institute of Macromolecular Chemistry ASCR.
doxorubicin is released inside the cell and then acts analogously
to free doxorubicin. Controlled release of doxorubicin within
cancer cells is achieved by hydrolysis of hydrazone conjugates
having doxorubicin bound via pH-sensitive bonds (6). Those
bonds are stable at physiological pH 7.4, but are effectively
degraded at pH 5. Inside the cell, doxorubicin is released quickly
in endosomes or lysosomes. Conjugates with doxorubicin bound
to the HPMA carrier through a hydrazone bond are highly
cytotoxic and cytostatic in vitro and in vivo (7).
On the other hand, intracellular events triggered by HPMA
copolymer-conjugated doxorubicin via an enzymatically degradable bond (PK1) remain unclear. These conjugates were
originally designed to release the drug from the oligopeptide
side chain by the activity of lysosomal proteases (8). Some
authors initially demonstrated that HPMA copolymer conjugates
with enzymatically degradable bonds can be stronger proapoptotic inducers than free doxorubicin (9). However, later it
was shown that both apoptosis and necrosis are induced in cells
exposed to proteolytically cleavable HPMA copolymer conjugates (10-12). Our recent data do not support the idea that the
enzymatic degradation and doxorubicin release from PK1 are
the most important mechanisms responsible for target cell death.
They rather suggest that immediately after penetrating the
plasma membrane the whole conjugate intercalates into the
membrane system of the cell, causing its collapse and death
(10, 11).
PK1 with enzymatically degradable bond between the oligopeptide spacer GFLG and doxorubicin was tested in phase I
10.1021/bc060246e CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/03/2007
Doxorubicin and HPMA Conjugates in Cancer Cells
and was recommended for phase II of clinical study (13). A
conjugate similar to PK1 but containing human nonspecific
antibody was used for the treatment of patients with advanced
cancer (14).
In this study, we used the EL-4 T-cell lymphoma to determine
the differences in intracellular action of free doxorubicin, PK1,
and HYD. We determined their impact on intracellular signaling pathways (mitogen-activated protein kinases, NFκB, and
apoptosis) and expression of genes involved in the regulation
of cell cycle, and showed differences in their intracellular
distribution.
EXPERIMENTAL PROCEDURES
Materials. Hydrazine monohydrate, methacryloyl chloride,
1-aminopropan-2-ol, 4-nitrophenol, 6-aminohexanoic acid, glycyl-L-phenylalanine, L-leucylglycine, 2,2′-azobis(isobutyronitrile) (AIBN), 2-(dimethylamino)ethyl methacrylate, dimethylformamide (DMF), N,N′-dicyclohexylcarbodiimide (DCC),
1-hydroxybenzotriazole (HOBT), dimethyl sulfoxide (DMSO),
tert-butylcarbazate, and doxorubicin hydrochloride (Dox.HCl)
were purchased from Fluka Chemie AG. 2,4,6-Trinitrobenzene1-sulfonic acid was purchased from SERVA Feinbiochemica
Heidelberg. All other reagents and solvents were of analytical
grade.
Synthesis and Characterization of Monomers. N-(2-Hydroxypropyl)methacrylamide (HPMA) was synthesized as described in (15) using Na2CO3 as a base in the methacryloylation
reaction (mp 70 °C; elemental analysis (calcd/found): 58.80/
58.98% C; 9.16/9.18% H; 9.79/9.82% N). N-(tert-Butoxycarbonyl)-N′-(6-methacrylamidohexanoyl)hydrazine (Ma-AH-NHNHBoc) was prepared by a two-step synthesis consisting of
methacryloylation of 6-aminohexanoic acid followed by reaction
of the resulting N-methacryloyl-6-aminohexanoic acid with tertbutylcarbazate as described in (7). Yield: 6.06 g (46%). Mp
110-114 °C. Elemental analysis: calcd C 57.70%, H 8.33%,
N 13.46%; found C 58.66%, H 8.84%, N 13.16%.
N-Methacryloylglycyl-DL-phenylalanyl-L-leucylglycine and
N-methacryloylglycyl-DL-phenylalanyl-L-leucylglycine 4-nitrophenyl ester (Ma-GFLG-ONp) were synthesized as described
earlier (1). Mp 134-136 °C. Amino acid analysis: Gly/L-Phe/
D-Phe/L-Leu ) 2.05/0.54/0.47/1.00. Elemental analysis (calcd/
found): C 59.89/59.21%; H 6.07/6.25%; N 12.04/12.32%.
HPLC showed two peaks of equal areas at 14.41 min (L-Phe
peptide) and 14.71 min (D-Phe peptide).
Purity of all monomers was examined by HPLC (LDC
Analytical, U.S.A.) using a reversed-phase column Tessek SGX
C18 (15 × 33 mm) with UV detection at 230 nm, eluent watermethanol with gradient 50-100 vol % methanol, flow rate 0.5
mL/min.
Synthesis of Polymer Precursors. Copolymer poly(HPMAco-MA-AH-NHNH2). Copolymer poly(HPMA-co-MA-AHNHNH-Boc) was prepared by radical solution polymerization
(AIBN, 1 wt %; monomer concentration, 14 wt % in methanol
solution; molar ratio HPMA/MA-AH-NHNH-Boc 93/7; 60 °C;
23 h) in a sealed ampule under nitrogen. The copolymer was
isolated by precipitation into a mixture of acetone/diethyl ether
(2:1). The protecting Boc group was removed by dissolution
of the polymer in concd trifluoroacetic acid (TFE), and a final
polymer was isolated after dilution with methanol by precipitation into ethyl acetate and purified by gel filtration using a
column filled with Sephadex LH-20 and methanol as a solvent.
The poly(HPMA-co-MA-AH-NHNH2) was isolated by precipitation into ethyl acetate, separated by filtration, and dried in
vacuum to constant weight.
Poly(HPMA-co- Ma-GFLG-ONp) was prepared by radical
precipitation polymerization as described in (15).
Bioconjugate Chem., Vol. 18, No. 3, 2007 895
HPMA homopolymer poly(HPMA) was prepared by radical
solution polymerization using AIBN as initiator (0.8 wt %) and
monomer concentration 16 wt % in methanol. Polymerization
was carried out at 60 °C for 22 h in a sealed ampule under
nitrogen; the polymer was isolated by precipitation into acetone
and reprecipitated from methanol into acetone.
Synthesis and Characterization of Polymer-Dox Conjugates. The conjugate bearing Dox attached via a pH-sensitive
hydrazone bond poly(HPMA-co-MA-AH-NHN ) Dox) (HYD)
and the conjugate bearing Dox attached via the enzymatically
degradable oligopeptide sequence poly(HPMA-co- Ma-GFLGDox) (PK1) were prepared as described earlier, HYD in (6) and
PK1 in (15).
The polymer-drug conjugates were freed of low-molecularweight impurities (such as Dox, 4-nitrophenol) by gel filtration
using a Sephadex LH-20 column with methanol elution and
tested for the content of the free drug using a Pharmacia FPLC
equipped with a Superose 6 column and by HPLC after
extraction of free Dox from an aqueous polymer solution into
chloroform. HPLC analysis of monomers was conducted on an
HPLC analyzer (LDC Analytical, U.S.A.) using a reversed-phase
column Tessek SGXC18 (125 × 4 mm) with UV detection at
230 nm, solvent methanol-water, gradient 50-100% methanol,
and flow rate 0.5 mL/min. Content of Dox in the conjugates
was measured spectrophotometrically in water using ) 11
500 L mol-1 cm-1 at λ ) 488 nm).
Amino acid analysis was performed using an amino acid
analyzer (LDC Analytical, U.S.A.) (precolumn OPA derivatization, reversed-phase column Tessek SGX C18, 250 × 4 mm,
gradient sodium acetate buffer-methanol, fluorescence detector
Fluoromonitor 4100).
Determination of the molecular weight of all polymers was
carried out with a FPLC Pharmacia system equipped with RI,
UV, and multiangle light scattering DAWN DSP-F (Wyatt Co.,
U.S.A.) detectors using 0.3 M acetate buffer pH 6.5 and a
Superose 6 column.
The content of hydrazide groups in a precursor and HYD
conjugate was determined by a modified TNBSA assay as
described (16).
The following polymers were prepared: HPMA copolymer
bearing Dox attached via enzymatically degradable spacer
(PK1), Mw ) 24.8 kDa, Mw/Mn ) 1.5, total Dox content 5.2 wt
%, free Dox content <0.1%. HPMA copolymer bearing Dox
attached via pH-sensitive hydrazone bond (HYD), Mw ) 25
kDa, Mw/Mn ) 1.8, total Dox content 5.2 wt %, free Dox content
<0.3%.
Detection of MAPK Phosporylation by Immunoblotting.
1-5 × 107 EL-4 cells were incubated with tested samples at a
concentration corresponding to 3, 10, or 50 × IC50 of doxorubicin equivalent. After incubation periods of 5, 15, 30, 60, or
120 min, the media from the cell culture were aspirated, and
cells were washed twice with an ice-cold Tris-buffered saline
(TBS) with 1 µM Na3VO4 (phosphatase inhibitor) and centrifuged (4000 g, 4 °C). Then, EL-4 cells were resuspended in
Extract buffer composed of 1% Nonidet P-40 (Pierce), 1 mM
Na3VO4, 1 mM EDTA, 2 mM EGTA, 10 mM NaF, 1 mM DTT,
5% Protease mix (Sigma), 1 mM PMSF phenylmethylsulfonyl
fluoride (PMSF), and TBS of pH 7.4, and passed ten times
through a syringe needle (25-30G). After 60 min of incubation
at 4 °C, cell lysates were centrifuged at 14 000 g (4 °C), and
protein concentration in aspirated supernatants was determined
by the Bradford method. 20 µg protein (final protein load per
lane) aliquots were mixed 2:1 with 3× sample buffer composed
of 62.5 mM Tris, 30% glycerol, 20.8 µM SDS, 50 mM DTT,
and 0.03% w/v bromophenol blue and stored at -20 °C until
needed. Then, standard electrophoresis using 10% polyacrylamide gel followed by standard blotting semidry procedure with
896 Bioconjugate Chem., Vol. 18, No. 3, 2007
nitrocellulose membrane was performed. 5 µL of a biotinylated
protein ladder (Cell Signaling, MA) was used to estimate the
molecular weight of protein bands. The membrane was washed
in 25 mL of TBS and blocked for 60 min in 5% low-fat milk
or 2% BSA in TBS at room temperature before overnight
incubation with 1:1000 primary rabbit-anti-mouse phospho-p38,
phospho-SAPK/JNK, or anti-mouse phospho-ERK antibodies
(Cell Signaling, MA) at 4 °C in TBS with 0.1% Tween 20 and
5% BSA. Proper washing with TBS containing 0.1% Tween
20 (TBS/T) preceded 60-120 min of incubation with anti-rabbit
or anti-biotin, HRP-linked IgG (1:1000) (Cell Signaling, MA)
in 5% low-fat milk or 2% BSA in TBS at room temperature.
Finally, the membranes were washed in TBS/T, developed by
chemiluminescence reaction and exposed to LAS-1000 CCD
camera (Fujifilm). Then, the Aida Image Analyzer v 3.2.8.
(Raytest, Germany) was used for data quantification.
The molecular weights of the respective kinases were used
as internal standards of assay specificity. The molecular weights
of detected phospho-proteins were determined by using protein
markers, compared with the data provided by the manufacturer
and found to be identical with respective MAPKs. Untreated
cells were used as another internal control. The level of
respective phosphorylated MAPK in the untreated cell was
regarded as a reference value to which all other values were
referred.
In Vitro Proliferation Assay. An EL-4 murine T-cell
lymphoma cell line was grown in cultivation flasks at 37 °C
with 5% CO2 in RPMI 1640 medium supplemented with heatinactivated 10% fetal calf serum (FCS), 2 mM L-glutamine, 50
mM 2-mercaptoethanol, 4.5 g/L glucose, 1 mM sodium pyruvate, 100 U/mL penicillin, and 100 µg/mL streptomycin. 2 ×
105 cells/mL were cultured in Nunc 96-well flat-bottom plates
with or without Dox.HCl, HYD, or PK1, and/or MAPK
inhibitors. In order to exclude any effect of tested agents on
noninhibited kinase pathways, EL-4 cells were pretreated with
SP600125 (JNK inhibitor), SB203580 (p38 inhibitor), or with
U0126 (ERK inhibitor) (Sigma) MAPK inhibitors for 30 min
prior to addition of Dox, PK1, or HYD in 48 h proliferation
assays. In short-term assays, which were performed in order to
avoid indirect effects such as changes in gene expression that
could happen during the 48 h inhibition, inhibitors were added
30 min prior to [3H]thymidine addition in 5 h MAPK inhibition
assays. All samples were tested in pentaplets. The plates were
then cultured in 5% CO2 for 48 h at 37 °C. 18.5 kBq of [3H]thymidine in 50 µL was added per well for the last 5 h of
incubation. The cells were then harvested (Tomtec, Orange, CT)
onto glass filter mats. After drying, the fiber filter was placed
into a sample bag, a solid scintillator MeltiLex A (Wallac, Turcu,
Finland) was applied, and the bags were sealed (Microsealer,
Wallac). Counting was performed in a 1450 MicroBeta TriLux
scintillation counter (Wallac).
Flow Cytometry. Control, Dox, HYD, or PK1 treated cells
were incubated for 24 or 48 h with or without MAPK inhibitors,
as described in the In Vitro Proliferation Assay section. To
exclude any unspecific effect of tested agents on noninhibited
kinase pathways, kinase inhibitors were added 30 min prior to
Dox, HYD, or PK1 treatment. All agents were used at a
concentration of 1 × IC50. After incubation, cells were washed
and collected at 4 °C in PBS. Then, 0.01 µg/mL of Hoechst
33258 (Molecular Probes, Oregon) in 10 µL was added. After
20 min of incubation at 4 °C in the dark, 50 µL of ice-cold
PBS was added, and at least 50 000 cells were analyzed by
FACS (LSRII, BD, San Jose, CA) with the use of FowJo
software (Tree Star, San Carlos, CA).
Detection of Caspase 3 Activity. 5 × 106 EL-4 cells were
incubated with tested samples at a concentration corresponding
to 3 × IC50 for different time intervals. After the incubation
Kovář et al.
periods, the cells were washed twice with PBS, and the pellets
were resuspended in 50 µL of Lysis buffer (Caspase 3 Assay
Kit, Sigma). Cells were lysed on ice for 20 min, and 50 µL of
supernatant was collected after centrifugation (12 000 rpm, 5
min, 4 °C). The Bradford method was used to determine protein
concentration in cell lysates. The activity in cell lysates was
measured spectrophotometrically at 405 nm using the Caspase
3 Assay Kit (Sigma). Caspase 3 activity was finally expressed
as the amount of p-nitroaniline (nmol) released from synthetic
Caspase 3 substrate/µg of total protein in cell lysate.
Binding Activity of NFKB. 5 × 106 EL-4 cells were
incubated with tested samples at a concentration corresponding
to 10 × IC50 for different time intervals. After the incubation
period, the cells were washed twice with PBS, and the pellets
were resuspended in Lysis buffer (Nuclear Extraction Kit,
Chemicon). Cells were lysed on ice and passed five times
through syringe with small (25-30) gauge needle. After
centrifugation at 8 000 g for 20 min at 4 °C, the pellets were
resuspended in nuclear extraction buffer, and nuclei were
disrupted by lysis with the use of a 27G needle. After
centrifugation (16 000 g, 5 min, 4 °C), protein concentration
was determined with the use of the Bradford method, and
aliquots were frozen at -80 °C until needed. Then, the binding
assay was performed with the use of the NFκB p50/p65 Assay
Kit (Chemicon). Briefly, in the streptavidin-coated 96-well plate,
the biotinylated DNA binding sequence for NFκB was mixed
with nuclear extract. After addition of specific antibodies, the
amount of p50 or p65 subunits was detected spectrophotometrically at 450/650 nm. The NFκB binding activity was finally
expressed as percent of controls.
Intracellular Localization of Doxorubicin. 1 × 106 cells
were incubated for 24 h with 5 µg/mL of free doxorubicin or
an adequate equivalent contained in HYD conjugate. For PK1
conjugate, a concentration corresponding to 20 µg of doxorubicin per mL was used. The excitation of doxorubicin at 488
nm and emission at 560 nm was used to visualize the
intracellular localization of doxorubicin under Leica TCS-SP2
Multiphoton Confocal Laser Scanning Microscope (TCS-MP).
Intensity profiles were performed by using AnalySIS Software
(SIS software, Germany).
RNase Protection Assay. Total RNA was extracted from
EL-4 cells exposed to 3 × IC50 for different time intervals using
the RNA blue reagent (Top-Bio, Czech Republic). Total RNA
was quantified specrophotometrically and dried by vacuum
centrifugation for 1 h. To determine the levels of cytokine
mRNA transcripts, a RiboQuant Multi-probe RNase Protection
Assay System (Pharmingen) was used following the manufacturer’s instruction. R[32P] UTP (Izotop, Budapest, Hungary) was
used for the synthesis of antisense RNA probe sets. 32P-labeled
protected fragments were resolved by electrophoresis on 5%
long ranger gel (BMA, Rockland, Maine). Gels were dried and
radioactivity visualized by phosphorimaging. The level of
mRNA was evaluated for the genes involved in cell cycle
regulation (cyclin A1, A2, B1, B2, C, D1, D2, D3, sin3, c-, n-,
l-, b-myc, max, mad, mxi, mad3, mad4, mnt) and genes involved
in apoptosis signaling (caspase 8, fasL, fas, fadd, fap, faf, trail,
TNFRp55, tradd, rip, bcl-W, bfl-1, bcl-x, bak, bax, bcl-2, bad).
For each individual sample including the control, mRNA levels
for the genes listed above were expressed in % of mRNA level
for the L32 housekeeping gene. Final results are presented as a
ratio of mRNA levels for the particular gene in treated cells
and nontreated controls.
Statistical Analysis. The significance of any differences
obtained between experimental groups was evaluated by the
Student’s t-test. In all statistical analyses, significance was
assessed at the P e 0.05 level.
Bioconjugate Chem., Vol. 18, No. 3, 2007 897
Doxorubicin and HPMA Conjugates in Cancer Cells
Table 1. Cytostatic Effects of Free Doxorubicin, HPMA-Bound
Doxorubicin, and MAPK Inhibitorsa
sample
IC50 concentration (µM) ( S.D.
Dox.HCl
HYD
PK1
SP600125
SB203580
U0126
0.014 ( 0.001
0.066 ( 0.013
45.970 ( 3.844
6.36 ( 0.921
8.12 ( 0.725
38.00 ( 3.453
a IC
50 values determined in EL-4 cell line for free doxorubicin, HYD,
PK1 conjugate, and JNK inhibitor SP600125, p38 inhibitor SB203580, and
inhibitor of MEK/ERK U0126.
Figure 1. Phosphorylation of MAPKs in EL-4 cells treated with
different forms of doxorubicin. Phosphorylation of JNK (a), p38 (b),
and ERK (c) MAP kinases in lysates of EL-4 cells after 5 min (empty
bars), 30 min (dotted bars), or 120 min (full bars) treatment with
different forms of doxorubicin (50 × IC50) relative to control cells
(100%). Representative western blot bands are shown for each sample
below the corresponding column. Dox.HCl, free doxorubicin; HYD,
doxorubicin bound to HPMA copolymer via hydrolytically degradable
bond; PK1, doxorubicin bound to HPMA copolymer via enzymatically
cleavable bond. *The difference against control was significant at P <
0.05 (Student’s t-test).
RESULTS
MAPKs Phosphorylation. In preliminary studies, we used
the concentrations of 3×, 10×, and 50× IC50 of Dox, HYD,
and PK1 on MAPK phosphorylation (not shown). The highest
phosphorylation rates were achieved with the highest concentration of 50 × IC50, and it was used in assays with phosphorylated
JNK (Figure 1a), p38 (Figure 1b), and ERK (Figure 1c).
Incubation of EL-4 cells with free doxorubicin stimulated
phosphorylation of p38 kinase up to 450% of control, whereas
it inhibited phosphorylation of JNK and ERK almost by 50%
in comparison with control cells. The maximum stimulation of
p38 phosphorylation was achieved during the longest incubation
lasting 120 min, and inhibition of JNK and ERK was acheived
during both the 30 and 120 min incubation periods. Hydrazone
conjugate (HYD) stimulated the phosphorylation of the stressactivated protein kinase JNK up to 360% of controls in the 2 h
incubation period. ERK phosphorylation was inhibited after both
the 30 and 120 min incubation periods with HYD, whereas a
slight inhibition of p38 phosphorylation turned into minute
enhancement during the long-term incubation. Phosphorylation
of ERK was enhanced up to 205% of controls after incubation
with PK1 conjugate. PK1 achieved an almost total (95%)
inhibition of JNK phosphorylation. During incubation of EL-4
cells with PK1 conjugate, previous almost absolute inhibition
of p38 phosphorylation turned into the same level of phoshorylation as that of the control after longer incubation.
EL-4 Cell Proliferation Assay and IC50 Values. IC50 values
summarized in Table 1 represent the concentration of doxorubicin or MAPK inhibitors required for 50% inhibition of EL-4
cell proliferation after 48 h. Free doxorubicin exhibited the
highest cytostatic effect at 0.014 µM, IC50 value of hydrazone
conjugate was 0.066 µM (doxorubicin equivalent), while PK1
conjugate showed the lowest cytostatic effect of 45.97 µM. IC50
of SP600125, SB203580, and U0126 were 6.36 µM, 8.12 µM,
and 38 µM, respectively.
After 48 h of incubation with 1 × IC50 value of SP600125
(JNK pathway inhibitor), EL-4 cells proliferated about 50% less
when coincubated with or without hydrazone or PK1. On the
other hand, when coincubated with free doxorubicin, EL-4 cells
proliferated twice more in comparison with control cells
incubated only with free doxorubicin without SP600125. EL-4
cells incubated for 48 h with free doxorubicin or HYD conjugate
followed by 5 h of JNK inhibition with SP600125 proliferated
about 50% less than cells without SP600125 inhibitor with free
doxorubicin or HYD conjugate. This short-term, 5 h JNK
pathway inhibition did not have any effect on proliferation of
control or PK1-treated cells (Figure 2a).
Neither 5 (not shown) nor 48 h inhibition of both the p38
pathway by SB203580 (Figure 2b) and ERK pathway by U0126
(Figure 2c) caused any significant differences in the proliferation
of controls or cells treated with doxorubicin, HYD, or PK1. In
those incubation assays, proliferation of all cells was inhibited
approximately by 50% relative to corresponding untreated
control cells.
FACS Analysis of Cell Viability. Inhibition of the JNK
pathway by SP600125 (Figure 3a) for 48 h had no significant
effect on cellular viability in controls, HYD-, or PK1-treated
cells. Coincubation of EL-4 cells with doxorubicin and SP600125
reduced cellular viability up to 86% relative to the cells with
uninhibited JNK pathway, whereas cells exposed only to
Dox.HCl reduced their viability by approximately up to 60%.
On the other hand, inhibition of the JNK pathway had no
significant effect on viability of cells exposed to HYD or PK1
conjugates. A 24 h inhibition of the JNK pathway did not
significantly affect the number of dead cells in controls or any
tested samples.
898 Bioconjugate Chem., Vol. 18, No. 3, 2007
Kovář et al.
Figure 2. Proliferative activity of EL-4 cells incubated with different
forms of doxorubicin and with/without MAPK inhibitors. Proliferation
of EL-4 cells after a 48 h treatment with different forms of doxorubicin
(1 × IC50) relative to control cells (100%). Dox.HCl, free doxorubicin;
HYD, doxorubicin bound to HPMA copolymer via hydrolytically
degradable bond; PK1, doxorubicin bound to HPMA copolymer via
proteolytically cleavable bond. Cells were coincubated either for 5 h
(empty bars) or for 48 h (full bars) with MAPKs inhibitors (1 × IC50)
or without MAPK inhibitors (dotted bars). Inhibition of JNK (a), p38
(b), or inhibition of ERK (c). *The difference against the group without
inhibitor but exposed to the same form of doxorubicin was significant
at P < 0.05 (Student’s t-test).
Figure 3. Viability of EL-4 cells incubated with different forms of
doxorubicin and with/without MAPK inhibitors. Viability of EL-4 cells
after a 48 h treatment with different forms of doxorubicin (1 × IC50)
relative to control cells (100%). Dox.HCl, free doxorubicin; HYD,
doxorubicin bound to HPMA copolymer via hydrolytically degradable
bond; PK1, doxorubicin bound to HPMA copolymer via enzymatically
cleavable bond. Cells were coincubated either for 24 h (empty bars) or
for 48 h (full bars) with MAPKs inhibitors (1 × IC50) (a) inhibiting
JNK, (b) p38, or (c) inhibiting ERK. Dotted bars show cells without
inhibited MAPK pathway incubated with different forms of doxorubicin.
*The difference against the group without inhibitor, but exposed to
the same form of doxorubicin was significant at P < 0.05 (Student’s
t-test).
Inhibition of the p38 kinase pathway (Figure 3b) for 24 or
48 h did not significantly affect the viability of control cells
nor of those treated with Dox.HCl, HYD, and PK1 conjugate
when compared to cells exposed to free doxorubicin or HPMA
conjugates without inhibitor.
U0126 inibitor had been shown to have toxic effects even
on control EL-4 cells incubated in cultivation media only.
Coincubation with free doxorubicin or HPMA conjugate
enhanced U0126 toxicity. Cells exposed to Dox.HCl, HYD, or
PK1 conjugate showed reduced viability by 94%, 95%, and
89%, respectively, in 48 h incubation. 24 h incubation showed
85%, 88%, and 68% viability reduction in Dox.HCl, HYD, and
PK1 treated cells, respectively (Figure 3c).
Caspase 3 Activity. Caspase 3 was rapidly activated in EL-4
cells after treatment with free doxorubicin or hydrazone
conjugate, and its activity increased more than 7- and 5-fold
relative to controls, respectively. PK1 conjugate did not enhance
the activity of this effector caspase even over an incubation
period as long as 30 h (Figure 4a).
NFKB Binding Activity. The NFκB p50 binding activity was
significantly reduced in cell lysates after treatment with free
doxorubicin and hydrazone conjugate at all tested time intervals.
The lowest binding was detected in the longest 16 h treatment
and was as low as 43% with Dox.HCl and 42% with hydrazone
conjugate, as compared with controls. On the other hand, in
PK1-treated cells, the p50 binding was significantly reduced
only after 4 and 8 h incubation (Figure 4b). The binding activity
of the p65 subunit was not significantly changed either with
Dox.HCl and with hydrazone, or in PK1-treated cells (data not
shown).
Intracellular Localization. The confocal laser scanning
microscopy visualizing the intracellular localization of free or
HPMA-conjugated doxorubicin in native cells showed differences in the intracellular fate of these agents. The fluorescence
of free doxorubicin was localized mainly inside the nucleus and
endosomes, whereas the fluorescence of doxorubicin in the PK1
conjugate was clearly localized in cell membranes (cytoplasmic,
nuclear, endosomal, Golgi apparatus, etc.). The doxorubicin
contained in the hydrazone conjugate was shown to be localized
in endosomes and lysosomes, and released doxorubicin was
detected inside the nucleus (Figure 4c).
Bioconjugate Chem., Vol. 18, No. 3, 2007 899
Doxorubicin and HPMA Conjugates in Cancer Cells
Expression of Genes Involved in the Regulation of Cell
Cycle and Apoptosis. Expression of selected mRNA for genes
involved in the regulation of cell cycle and apoptosis was
determined in EL-4 cells treated with 3 × IC50 values of tested
conjugates. The results for the most important genes with
significantly altered expression are shown in Figure 5. Expression of an important proapoptotic gene bad was elevated up to
112% and 127% after incubation with free doxorubicin and the
hydrazone conjugate, respectively, whereas the PK1 conjugate
significantly reduced the level of mRNA for this gene to 48%,
as compared with controls. Expression of another gene involved
in apoptosis signaling - fas, were elevated after incubation with
all tested forms of doxorubicin. The enhancement caused with
the PK1 conjugate is significantly lower and more timedependent in comparison with free doxorubicin or hydrazone
conjugate. We did not observe any changes in the expression
of bcl-2 family members tested, nor in the expression of bax
and bak.
PK1-treated cells exhibited significantly increased levels of
c-myc and max that are important for cell cycle progression.
This nearly 150% and 120% increase of c-myc and max,
respectively, turned into inhibition later on. Free doxorubicin
and HYD conjugate quickly and strongly reduced the amount
of those two genes.
The expression of all tested cyclines was quickly reduced
after incubation with Dox.HCl and HYD conjugated, but this
inhibition was slighter and slower after incubation of EL-4 cells
with PK1 conjugates. Cyclines A2 and D2 are shown as typical
indicators of action of tested forms of doxorubicin on expression
of cycline genes.
DISCUSSION
Figure 4. Caspase 3 activity, binding activity of p50 subunit of NFκB
and intracellular distribution in EL-4 cells after treatment with different
forms of doxorubicin. (a) The activity of caspase 3 in cells incubated
with different forms of doxorubicin (3 × IC50). Caspase 3 activity is
expressed as the amount of p-nitroaniline (nmol) released from synthetic
caspase 3 substrate/µg of total protein in cell lysate and are related to
control (control cells ) 100%). (b) Binding activity of p50 subunit of
NFκB in cells treated with 10 × IC50 of different forms of doxorubicin
for 4 (empty bars), 8 (dotted bars), 16 (hatched bars), or 24 (full bars)
hours relative to control cells (100%). *The difference against control
was significant at P < 0.05 (Student’s t-test). (c) Intracellular
localization of free doxorubicin, hydrazone conjugate, both with 5 µg
of doxorubicin per mL and with PK1 conjugate with 20 µg of
doxorubicin per mL in EL-4 cells after 24 h of treatment. Cross-section
lines represent the intensity profiles demonstrating the fluorescence of
doxorubicin. Free doxorubicin (-b-, Dox.HCl), doxorubicin bound to
HPMA copolymer via hydrolytically degradable bond (-9-, HYD),
doxorubicin bound to HPMA copolymer via proteolytically cleavable
bond (-2-, PK1).
The MAPK transfers signals from the cell membrane to the
nucleus by phosphorylation of transcription factors like Elk-1.
JNK, p38, and ERK are serine/threonine kinases that are
activated by dual phosphorylation of the threonine/tyrosine motif
by upstream kinases (17). Using specific antibodies, we found
that Dox.HCl, HYD, and PK1 conjugates activate different
MAPKs. Free doxorubicin and HYD activated the p38 and JNK,
respectively. These kinases are typically activated in response
to cell stress, DNA damage, and precede apoptotic cell death
(18). Moreover, both the free drug and HYD conjugate enhanced
the expression of proapoptotic gene bad and increased the
activity of caspase 3, an effector apoptotic caspase whose
activation is an irreversible step in apoptotic cell death. Both
the Dox.HCl and HYD conjugate also increased the expression
of Fas and thus made the cells more sensitive to FasL-mediated
apoptotic cell death. These two agents also inhibited the
activation of the p50 subunit of NFκB transcription factor that
provides antiapoptotic signals.
On the other hand, the PK1 conjugate activated the ERK that
is usually associated with cell growth and differentiation (17).
In contrast to free doxorubicin and HYD conjugate, the PK1
conjugate reduced the expression of bad, did not enhance the
activity of caspase 3 (up to 30 h of incubation), and was not
inhibitory for p50 activation in longer time intervals. Surprisingly, neither free drug nor HPMA conjugates changed the
expression of antiapoptotic genes as well as the expression of
cyclin-dependent kinases (data not shown).
The mode of action of the HYD conjugate resembles that of
free doxorubicin, as doxorubicin is released from the HYD
conjugate by hydrolysis, a relatively fast process, and it acts as
free doxorubicin by triggering apoptosis. This explains the
similar mode of action of these agents in our study that has
been demonstrated previously with another cell line (5). In
contrast, the mechanism of action of the PK1 conjugate is less
clear. Recently, we have shown new data which do not support
the enzymatic degradation of PK1 conjugate in vivo and thus
900 Bioconjugate Chem., Vol. 18, No. 3, 2007
Kovář et al.
Figure 5. Expression of selected genes on mRNA level in EL-4 cells treated with different forms of doxorubicin. The level of mRNA for selected
genes determined by RNAse protection assay in the EL-4 cells after treatment with different forms of doxorubicin (3 × IC50). For each individual
sample including control, mRNA levels were expressed in % of mRNA level for L32 housekeeping gene. Final results are presented as a ratio of
mRNA levels for the particular gene in treated cells and non-treated controls. Free doxorubicin (-b-), doxorubicin bound to HPMA copolymer via
hydrolytically degradable bond (-9-), doxorubicin bound to HPMA copolymer via proteolytically cleavable bond (-2-). *The difference against the
free doxorubicin and hydrazone-treated groups was significant at P < 0.05. †The difference against the free doxorubicin and hydrazone (except for
the 4 h incubation period with hydrazone) treated groups was significant at P < 0.05. ‡The difference against the free doxorubicin and hydrazone
(except for the 4 h incubation period with both free doxorubicin and hydrazone) treated groups was significant at P < 0.05 (Student’s t-test).
show that the release of free doxorubicin is not a dominant
mechanism (15). Instead, we suggest that the whole PK1
conjugate interacts with cell membranes, causing their damage
(changed permeability, blockade of intracellular trafficking, etc.)
and subsequently resulting in cell death (10, 11). These data
are supported by our recent finding that both the Dox.HCl and
HYD conjugate trigger stress-activated signaling pathways JNK
and p38 and strongly induce apoptosis. On the other hand, the
PK1 conjugate activates ERK and suppresses the activation of
JNK. Thus, PK1 shows an opposite pattern in activation of
different MAPKs in comparison to Dox.HCl and HYD.
The data showing the effect of free or HPMA-conjugated
doxorubicin on expression of genes involved in cell cycle
(c-myc, max, and cyclines) demonstrate that Dox.HCl and HYD
conjugate inhibit cell proliferation quickly, whereas the PK1
conjugate does not inhibit the cell cycle efficiently within the
first 24 h.
Further, we used inhibitors specific for JNK, p38, and ERK
to evaluate their impact on proliferation and cell death. These
inhibitors alone showed some inhibition of proliferation which
is shown in Table 1. Doxorubicin-treated cells with inhibited
JNK proliferated more than twice as much as the cells without
JNK inhibitor. Moreover, the viability of the cells with inhibited
JNK was higher than that of cells treated only with free
doxorubicin, which clearly shows that JNK plays a key role in
mediating the cytostatic and cytotoxic effect of Dox.HCl. On
the other hand, the death of control cells and HPMA conjugate
treated cells was not significantly changed after treatment with
the JNK inhibitor. It seems that, in doxorubicin treated cells,
the JNK pathway triggers a lethal, probably proapoptotic signal,
and blocking of this signal protects cells against doxorubicininduced cell death (19). In contrast, this important role of the
JNK pathway was not seen in cells treated with PK1 and even
with HYD conjugate. This is surprising, because the HYD
conjugate has been reported to possess properties very similar
to the free drug (5). As shown previously with different cell
lines (10, 11), the intracellular distribution of the PK1 conjugate
greatly differs from that of free doxorubicin and the hydrazone
Doxorubicin and HPMA Conjugates in Cancer Cells
conjugate. These data discussed previously support our present
results and show one of the possible mechanisms of action.
Furthermore, we used a confocal laser scanning microscopy
to visualize the intracellular localization of doxorubicin in living
cells exposed to different forms of doxorubicin. We demonstrated significant differences in the intracellular fate of these
agents. The cells treated with free doxorubicin showed localization of the drug mainly inside the nucleus and acidic organelles
(10). The fluorescence pattern within cells incubated with
hydrazone conjugate was similar to that of the free drug, and
the strong signal within the nucleus verified the effective release
of doxorubicin from its carrier. By contrast, the fluorescence
signal in cells treated with the PK1 conjugate was clearly
localized almost solely within cell membranes (cytoplasmic,
nuclear, endosomal, Golgi apparatus, etc.), and we were not
able to detect any nuclear fluorescence. The latter finding
indicates that PK1 conjugate was not intracellularly degraded
and doxorubicin was not released in sufficient amount to be
detected within cell nucleus. This seems to be in contradiction
to the data published by Duncan’s team (20), as well as two
other groups who demonstrated nuclear localization of doxorubicin presumably released from the PK1 conjugate (21, 22).
To put these conflicting data into accord, we propose that the
differences might be due to the use of different cancer cell lines
for the studies, as different cells have different trafficking
pathways and the level of enzymes capable of cleaving
doxorubicin from its carrier. Another explanation might be that
doxorubicin detected within the nucleus was due to a contamination of the PK1 conjugate by free doxorubicin rather than to
its intracellular release from the conjugate. This assumption is
based on the fact that Duncan’s group (20) reported that the
PK1 conjugate used in their study contained 0.39% of free
doxorubicin, while the content of free doxorubicin in our PK1
conjugate is up to the limit of 0.1% of total doxorubicin. In
fact, the content of free doxorubicin in the PK1 conjugate
synthesized by us is probably much lower, but the 0.1% content
of free doxorubicin corresponds to doxorubicin concentration
which is near the detection limit of the analytical assay.
Unfortunately, the two remaining papers (21, 22) demonstrating
the doxorubicin release from the PK1 conjugate do not even
indicate free doxorubicin content in their PK1 conjugates and
therefore neither disprove nor corroborate our hypothesis.
To sum up, we demonstrated that the conjugation of doxorubicin to HPMA copolymer carrier modulates the mechanism
of its intracellular action. As mentioned before (5), the intracellular action of hydrazone conjugate resembles that of free
doxorubicin, but in this study we have found different roles for
MAP kinases. The PK1 conjugate has different and in some
parameters opposite mechanisms of action from both free
doxorubicin and the hydrazone conjugate. Since PK1 triggers
different intracellular pathways involved in regulation of cell
death and/or proliferation than free doxorubicin, one can
hypothesize that there can also be a different spectrum of tumors
sensitive to PK1. As an example, Dox.HCl is not used for the
treatment of colorectal cancer at all, but PK1 showed some
therapeutic effect in patients with this type of cancer (13). If
such a hypothesis is shown to be true, the finding that the
cancerostatic activity of a particular cytostatic drug toward
different types of tumor can be modified by binding to a
polymeric carrier is both scientifically interesting and important
for the development of new polymer-bound drugs for treatment
of tumors resistant to this drug in the free form.
ACKNOWLEDGMENT
This research was supported by the Grant Agency of the
Czech Republic (grants 305/05/2268 and 310/03/H147), the
Bioconjugate Chem., Vol. 18, No. 3, 2007 901
grant of MEYS CR 1M0505, and by the Institutional Research
Concept AV0Z50200510.
LITERATURE CITED
(1) Rihova, B., Bilej, M., Vetvicka, V., Ulbrich, K., Strohalm, J.,
Kopecek, J., and Duncan, R. (1989) Biocompatibility of N-(2hydroxypropyl) methacrylamide copolymers containing adriamycin.
Immunogenicity, and effect on haematopoietic stem cells in bone
marrow in vivo and mouse splenocytes and human peripheral blood
lymphocytes in vitro. Biomaterials 10, 335-42.
(2) Maeda, H. (2001) The enhanced permeability and retention (EPR)
effect in tumor vasculature: the key role of tumor-selective
macromolecular drug targeting. AdV. Enzyme Regul. 41, 189-207.
(3) Rihova, B. (1995) Antibody-targeted polymer-bound drugs. Folia
Microbiol. (Prague, Czech Repub.) 40, 367-84.
(4) Minotti, G., Menna, P., Salvatorelli, E., Cairo, G., and Gianni, L.
(2004) Anthracyclines: molecular advances and pharmacologic
developments in antitumor activity and cardiotoxicity. Pharmacol.
ReV. 56, 185-229.
(5) Kovar, M., Kovar, L., Subr, V., Etrych, T., Ulbrich, K., Mrkvan,
T., Loucka, J., and Rihova, B. (2004) HPMA copolymers containing
doxorubicin bound by a proteolytically or hydrolytically cleavable
bond: comparison of biological properties in vitro. J. Controlled
Release 99, 301-14.
(6) Etrych, T., Chytil, P., Jelinkova, M., Rihova, B., and Ulbrich, K.
(2002) Synthesis of HPMA copolymers containing doxorubicin
bound via a hydrazone linkage. Effect of spacer on drug release
and in vitro cytotoxicity. Macromol. Biosci. 2, 43-52.
(7) Mrkvan, T., Sirova, M., Etrych, T., Chytil, P., Strohalm, J., Plocova,
D., Ulbrich, K., and Rihova, B. (2005) Chemotherapy based on
HPMA copolymer conjugates with pH-controlled release of doxorubicin triggers anti-tumor immunity. J. Controlled Release 110,
119-129.
(8) Kopecek, J., Rejmanova, P., Strohalm, J., Ulbrich, K., Rihova, B.,
Chytrý, V., Duncan, R., and Lloyd, J. B. (1991) Synthetic polymeric
drugs. U.S. Patent 5,037,883.
(9) Minko, T., Kopeckova, P., and Kopecek, J. (2001) Preliminary
evaluation of caspases-dependent apoptosis signaling pathways of
free and HPMA copolymer-bound doxorubicin in human ovarian
carcinoma cells. J. Controlled Release 71, 227-37.
(10) Hovorka, O., St’astny, M., Etrych, T., Subr, V., Strohalm, J.,
Ulbrich, K., and Rihova, B. (2002) Differences in the intracellular
fate of free and polymer-bound doxorubicin. J. Controlled Release
80, 101-17.
(11) Hovorka, O., Etrych, T., Šubr, V., Strohalm, J., Ulbrych, K., and
Řı́hová, B. (2005) HPMA based macromolecular therapeutics:
internalization, intracellular pathway and cell death depend on the
character of covalent bond between the drug and the peptidic spacer
and also on spacer composition. J. Drug Targeting 14, 391-403.
(12) Demoy, M., Minko, T., Kopeckova, P., and Kopecek, J. (2000)
Time- and concentration-dependent apoptosis and necrosis induced
by free and HPMA copolymer-bound doxorubicin in human ovarian
carcinoma cells. J. Controlled Release 69, 185-196.
(13) Vasey, P. A., Kaye, S. B., Morrison, R., Twelves, C., Wilson, P.,
Duncan, R., Thomson, A. H., Murray, L. S., Hilditch, T. E., Murray,
T., Burtles, S., Fraier, D., Frigerio, E., and Cassidy, J. (1999) Phase
I clinical and pharmacokinetic study of PK1 [N-(2-hydroxypropyl)methacrylamide copolymer doxorubicin]: first member of a new
class of chemotherapeutic agents-drug-polymer conjugates. Clin.
Cancer Res. 5, 83-94.
(14) Rihova, B. (2003) Antibody-targeted HPMA copolymer-bound
anthracycline antibiotics. Drugs Future 28, 1189-1210.
(15) Ulbrich, K., Subr, V., Strohalm, J., Plocova, D., Jelinkova, M.,
and Rihova, B. (2000) Polymeric drugs based on conjugates of
synthetic and natural macromolecules. I. Synthesis and physicochemical characterisation. J. Controlled Release 64, 63-79.
(16) Etrych, T., Jelinkova, M., Rihova, B., and Ulbrich, K. (2001) New
HPMA copolymers containing doxorubicin bound via pH-sensitive
linkage: synthesis and preliminary in vitro and in vivo biological
properties. J. Controlled Release 73, 89-102.
(17) Kyosseva, S. (2004) Mitogen-activated protein kinase signaling.
Int. ReV. Neurobiol. 59, 201-20.
(18) Wada, T., and Penninger, J. (2004) Mitogen-activated protein
kinases in apoptosis regulation. Oncogene 23, 2838-49.
902 Bioconjugate Chem., Vol. 18, No. 3, 2007
(19) Brantley-Finley, C., Lyle, C., Du, L., Goodwin, M., Hall, T.,
Szwedo, D., Kaushal, G., and Chambers, T. (2003) The JNK, ERK
and p53 pathways play distinct roles in apoptosis mediated by the
antitumor agents vinblastine, doxorubicin, and etoposide. Biochem.
Pharmacol. 66, 459-69.
(20) Seib, F. P., Jones, A. T., and Duncan, R. (2006) Establishment
of subcellular fractionation techniques to monitor the intracellular
fate of polymer therapeutics I. Differential centrifugation fractionation B16F10 cells and use to study the intracellular fate of HPMA
copolymer - doxorubicin. J. Drug Targeting 14, 375-90.
Kovář et al.
(21) Nan, A., Ghandehari, H., Hebert, C., Siavash, H., Nikitakis, N.,
Reynolds, M., and Sauk, J. J. (2005) Water-soluble polymers for
targeted drug delivery to human squamous carcinoma of head and
neck. J. Drug Targeting 13, 189-97.
(22) Omelyanenko, V., Kopeckova, P., Gentry, C., and Kopecek, J.
(1998) Targetable HPMA copolymer-adriamycin conjugates. Recognition, internalization, and subcellular fate. J. Controlled Release
53, 25-37.
BC060246E